NEW ISSUE (available from February 4, 2014)

HISTOLOGY AND HISTOPATHOLOGYCellular and Molecular Biology

Vol 29, issue 3, March 2014

8 A novel semi-quantitative method for measuring tissue bleedingG. Vukcevic, V. Volarevic, S. Raicevic, I. Tanaskovic, B. Milicic, T. Vulovic and S. Arsenijevicpp 353-360

 

Vukcevic G1, Volarevic V2, Raicevic S1, Tanaskovic I3*, Milicic B4, Vulovic T5 and Arsenijevic S6.

1Gynecologic and Obstetrics Clinic, Clinical Center of Montenegro, Montenegro.

2Department of Microbiology and Immunology, Faculty of Medical Sciences, University of Kragujevac, Serbia.

3Department of Histology and Embryology, Faculty of Medical Sciences, University of Kragujevac, Serbia.

4Department of Medical Statistics and Informatics, Faculty of Stomatology, University of Belgrade, Serbia.

5Center for Anesthesia, Clinical Center of Kragujevac, University of Kragujevac, Serbia.

6Gynecologic and Obstetrics Clinic, Clinical Center of Kragujevac, University of Kragujevac, Serbia.

*Corresponding author: Irena Tanaskovic, M.D., Ph.D.

Department of Histology and Embryology, Faculty of Medical Sciences, University of Kragujevac,

Serbia.

69 Svetozara Markovica Street, Kragujevac 34 000, Serbia.

Phone: + 38134306800

E-mail: irena.vuk@gmail.com

Source of Support: none

Running head: Semi-quantitative method for measuring tissue bleeding

Word count: 2712

Number of figures: 4

Number of tables: 3

Number of graphs: 2

Conflict of Interest: The authors state that they have no conflicts of interest.

Summary

In this study, we describe a new semi-quantitative method for measuring the extent of

bleeding in pathohistological tissue samples. To test our novel method, we recruited

120 female patients in their first trimester of pregnancy and divided them into three

groups of 40. Group I was the control group, in which no dilation was applied. Group

II was an experimental group, in which dilation was performed using classical

mechanical dilators. Group III was also an experimental group, in which dilation was

performed using a hydraulic dilator. Tissue samples were taken from the patients’

cervical canals using a Novak’s probe via energetic single-step curettage prior to any

dilation in Group I and after dilation in Groups II and III. After the tissue samples

were prepared, light microscopy was used to obtain microphotographs at 100x

magnification. The surfaces affected by bleeding were measured in the

microphotographs using the Autodesk AutoCAD 2009 program and its “polylines”

function. The lines were used to mark the area around the entire sample (marked A)

and to create “polyline” areas around each bleeding area on the sample (marked B).

The percentage of the total area affected by bleeding was calculated using the

formula: N = Bt x 100 / At where N is the percentage (%) of the tissue sample surface

affected by bleeding, At (A total) is the sum of the surfaces of all of the tissue samples

and Bt (B total) is the sum of all the surfaces affected by bleeding in all of the tissue

samples. This novel semi-quantitative method utilizes the Autodesk AutoCAD 2009

program, which is simple to use and widely available, thereby offering a new,

objective and precise approach to estimate the extent of bleeding in tissue samples.

Keywords: semi-quantitative method, bleeding extent, Autodesk AutoCAD, mechanical dilator, hydraulic dilator.

Introduction

The dilation of the cervical canal represents an invasive intervention (Bokstrom et al.,

1998; O’Connell et al., 2008). Mechanical instruments, such as the Hern, Hegar, Pratt,

Hanks and Denniston dilators, are used to sequentially dilate the cervix (Biron-

Shental et al., 2004; Gelber et al., 2006) by incrementally increasing the diameter of

the inserted dilator until the dilation procedure is complete. These mechanical dilators

require the use of appropriate force, which could permanently damage the cervical

tissue, thus rendering cervical dilation an invasive intervention with an associated

bleeding risk (Bokstrom et al., 1998; O’Connell et al., 2008). As a replacement for

mechanical dilators, we recently constructed a medical device for reliable mechanical

cervical dilation: a continuous controllable balloon dilator (CCBD) with the main

purpose of preventing uterine and cervical injury during cervical dilation (Arsenijevic

et al., 2012).

Although recent technical advancements have been directed toward developing novel

assays and methods for the detection and quantification of tissue injuries (Wang et al.,

2011; Ippolito et al., 2012; Volarevic et al., 2012), there are still no reports of reliable,

simple-to-use imaging software that could be used for the detection and quantification

of hemorrhage in tissue sections.

Herein, we present a novel semi-quantitative method for measuring the extent of

bleeding in cervical tissue samples obtained from patients who were dilated with

mechanical or hydraulic dilators. This method was implemented using Autodesk

AutoCAD 2009 software, a computer-aided design (CAD) program, which was used

primarily for preparing technical drawings in 2 dimensions (2-D).

CAD in conjunction with digitized anthropometric manikins was first used in the

medical field in 1990 for the analysis and control of stressful work postures (Ulin et

al., 1990). Recently, AutoCAD software was successfully used to analyze human

orbital asymmetry (Seiji et al., 2009) and to quantify the volume of mandibular ramus

and symphysis bone grafts before surgery (Verdugo et al., 2009; Verdugo et al.,

2010a), thus enhancing surgical planning prior to bone transplantation (Verdugo et al.,

2010b; Verdugo et al., 2012). Additionally, this software has proven to be reliable in

the semi-quantification of necrosis (Volarevic et al., 2009) and metastasis (Volarevic

et al., 2012) in experimental models of hepatitis and breast cancer.

Materials and methods

Study organization

We analyzed the bleeding extent in the pathohistological dilated cervical canal tissues

of patients who were part of a clinical study performed from 2010 to 2011 by the

Gynecology and Obstetrics Clinic of the Clinical Center of Montenegro (study no.

ISRCTN54007498).

An independent data and safety monitoring board of Gynecology and Obstetrics

Clinic of the Clinical Center of Montenegro monitored the study and reviewed the

protocol compliance and outcome data. The protocol was approved by the

institutional review board of each participating center.

Study patients

This study was comprised of 120 pregnant women that were randomly divided into

three groups. Group I (40 pregnant women) was the control group, in which no

dilation was performed. Group II (40 pregnant women) was an experimental group, in

which dilation was performed using a mechanical dilator. Group III (40 pregnant

women) was a second experimental group, in which dilation was performed using a

hydraulic dilator (Arsenijevic et al., 2004; Arsenijevic et al., 2012).

The patients were enrolled in the study using the following criteria: age between 19

and 40 years; pregnancy verified by an ultrasound; singleton pregnancy; gestational

age ≤10 months; uterus and cervix with normal findings; and absence of uterine

contractions or bleeding.

Patients were excluded from the study if any of the following criteria were met: any

signs of spontaneous abortion; any previous attempt at an abortion or use of

substances for cervical maturation; multiple pregnancy; the presence or, at minimum,

the suspicion of a septic abortion, followed by an elevated body temperature of 38°C

or higher, uterine pain and odorous vaginal secretions; the presence of any previous

intervention performed on the uterine cervix; uterine or cervical anomalies; an

intrauterine device in situ; or hemorrhagic and/or chronic diseases.

Patients from each group were matched for all potentially relevant factors that could

affect tissue bleeding: age, weight, smoking habits and medical history.

The experiments were undertaken with the understanding and appropriate informed

consent of each patient (Arsenijevic et al., 2012).

Pathohistological analysis

The pathohistological material was taken following the preparation of the patients as

if they were undergoing a routine gynecological intervention. The tissue samples were

taken from the cervical canal using a Novak’s probe via energetic single-step

curettage. After removing the instrument, the extracted narrow belt of the epithelium

was placed in a vessel with fixative solution.

To obtain relevant information and to determine the detailed structure for all three

subject groups, routine hematoxylin and eosin staining was performed according to

the standard protocol (Bancroft and Gamble, 2008).

The samples were fixed in a 4% buffered neutral formalin solution at room

temperature for 24 hours. After the fixation was complete, the samples were

dehydrated by dipping them into a graded alcohol series (70%, 96% and 100%),

followed by clearing in xylol and molding in paraplast.

The samples were cut to a width of 5 μm with a Leica SM 2000R Reinhart Austria

microtome. Staining was performed using Mayer’s hematoxylin followed by a 2%

eosin solution. Next, the samples were dehydrated, brightened and fixed to the plates

using Canada balsam.

The samples were analyzed under an Olympus BX51 light microscope.

Tissue bleeding was classified as no bleeding, intraepithelial bleeding, subepithelial

bleeding, stromal bleeding or combinations of the above.

Semi-quantitative method for determining the amount of bleeding

The amount of bleeding was examined by light microscopy with 100x magnification,

and pictures were obtained with a digital camera. The surface area affected by

bleeding was measured using the Autodesk AutoCAD 2009 software application for

design and graphics (Figures 1 and 2).

[Figures 1 and 2 here]

The procedure that was followed using the Autodesk AutoCAD 2009 program to

quantify tissue bleeding included several steps:

1. Each tissue sample area was visualized under a microscope, and the entire cervical

tissue sample was photographed.

2. Each tissue sample picture was individually stored in a newly opened Autodesk

AutoCAD 2009 file.

3. The “polylines” function of the program was used to draw “polyline” areas around

each sample (marked as A) and around each zone on the sample that was affected by

bleeding (marked as B).

4. The Autodesk AutoCAD 2009 program measured the surface of each “polyline”

area. First, it measured the surface of the “polyline” A area, and then it measured the

surface of each individual “polyline” B area.

5. The total area affected by bleeding was defined as a percentage using the following

formula:

 

N = Bt x 100 / At

where N is the percentage (%) of the tissue sample surface affected by bleeding

(Figures 3 and 4), At (A total) is the sum of the surfaces of all of the tissue samples

(At = A1 + A2 +…+ An, where n is the total number of photographs) and Bt (B total)

is the sum of all the surfaces affected by bleeding in all of the tissue samples (Bt = B1

+ B2 +…+ Bm, where m is the total number of marked surfaces affected by bleeding

in all of the photographs).

[Figures 3 and 4 here]

The percentage of the tissue surface that was affected by bleeding was calculated as

an objective measurement based on the appearance of bleeding in each tissue sample,

thereby avoiding the possibility of biased results due to the sample size.

Statistical analyses

The statistical analysis package SPSS 17.0 (IBM Corp., Armonk, NY, USA) was used

to perform the necessary statistical tests.

Descriptive data for all of the groups and variables are expressed as the mean

±standard deviation (SD), median, minimal and maximal values for continuous

measures or percent of a group for discrete measures.

Categorical data (number of patients with bleeding and different intensities of

bleeding) were analyzed using the Pearson chi-square test. A normal distribution of

continuous data (area of bleeding expressed in % and total sum of bleeding surface)

was analyzed using the Koglomorov-Smirnov test. All of our data were non-

parametric. The Kruskal-Wallis test was used for data analysis when more than two

groups were compared. The Mann-Whitney U test was performed for inter-group

comparisons.

The Spearman coefficient of correlation was calculated to show the reliability and

accuracy of our novel semi-quantitative methodology with the classical

pathohistology analysis.

All reported p-values were two sided. Differences were considered significant when

the p-value was <0.05.

Results

In this study, we assessed the effectiveness of different dilators by evaluating the

intensity of bleeding in cervical samples through classical pathohistological analysis

and a novel semi-quantitative method. Here, we first describe our evaluation of the

effectiveness of these different dilators by examining the intensity of bleeding using

classical pathohistological cervical biopsy. Next, we compare our new semi-

quantitative method with the pathohistological description of the bleeding

intensity. Finally, we evaluate the effectiveness of the different dilators using our

semi-quantitative method.

Using classical pathohistological cervical biopsy, we analyzed the impact of different

dilators on the occurrence of bleeding and demonstrated that there was a statistically

significant difference in the frequency of bleeding between the cervical samples from

the three groups of patients (χ2-test, p = 0.000) (Table 1). The minimum frequency of

bleeding was observed in the patients of Group I, and the highest frequency was

observed in the patients of Group II (Table 1).

[Table 1 here]

Using this classical method, we then analyzed the intensity of bleeding and found a

statistically significant difference between the cervical samples from the three patient

groups (χ2-test, p = 0.000) (Table 1).

This study describes a new method for the semi-quantitative estimation of the amount

of bleeding in a tissue sample. We believe that this new method is an adequate

replacement for the classical pathohistological analysis of bleeding in cervical

samples. The data obtained using this classical method are descriptive and are

categorized into different grade levels, as measured on an ordinal scale. Our novel

method of measuring the amount of bleeding involves the use of a numerical

measurement scale. To determine the reliability and validity of this new semi-

quantitative method, we compared the data resulting from the two methods using

Spearman's rank correlation coefficient; these results are shown in Table 2.

[Table 2 here]

The advantages of the new semi-quantitative method are reliability, objectiveness and

standardization of the parameters used to evaluate the soft tissue obtained from the

cervix.

Using our new method, we compared the bleeding area percentages in the tissue

samples for the patients with bleeding in Groups I, II and III and found a statistically

significant difference (Kruskal-Wallis test, p = 0.003) (Table 3).

[Table 3 here]

An inter-group comparison of the patients with bleeding confirmed that the difference

in the bleeding extent was statistically significant among the tissue samples taken from

the patients of Group I and Group II (Mann-Whitney U test; p = 0.009). A statistically

significant difference was also found between Group II and Group III (Mann-Whitney

U test; p = 0.007). However, the extent of the bleeding in the tissue samples from the

patients of Group I was not significantly different from the bleeding in tissue samples

from the patients in Group III (Mann-Whitney U test; p = 0.521).

By summing the surface area of the tissue samples affected by bleeding (among the

patients with bleeding) in all of the test groups, we demonstrated that the hydraulic

dilator was as efficient as the mechanical dilator but caused considerably less damage

to the cervical tissue, as shown in Graph 1.

[Graph 1 here]

The statistically significant differences between the three analyzed patient groups —

without dilation, after mechanical dilation and after hydraulic dilation — were also

calculated with regard to the percentages of the tissue sample surface areas affected

by bleeding (Kruskal-Wallis test; p = 0.003), as shown in Graph 2.

[Graph 2 here]

The largest percentage of surface bleeding was found in the group of women dilated

with a mechanical dilator (Group II). The percentage of bleeding in the group dilated

using a hydraulic dilator was much lower (Group III); this value was similar to the

percentage found for the tissues that were not dilated (Group I).

Discussion

Thus far, the literature has not described a method to precisely calculate the extent of

tissue bleeding. Although similar methods have been used in tissue engineering

(Verdugo et al., 2009; Verdugo et al., 2010a; Verdugo et al., 2012), immunology

(Volarevic et al., 2009), dentistry (Unal et al., 2009), neurology (Kofron et al., 2009),

anatomy (Seiji et al., 2009) and oncology (Volarevic et al., 2012), the method

described in this paper represents an original model for quantifying hemorrhaging in

target tissues. This model is applicable not only in gynecology (to estimate the extent

of bleeding in the cervical canal) but also to estimate hemorrhages in any organ.

Additionally, histological analyses of the cervical tissue in the first (Greer et al., 1992)

and third trimesters of pregnancy (Ichio et al., 1976) have rarely been conducted.

The primary advantages of this quantitative method over other descriptive methods

for hemorrhage analysis are its precision and objectivity. This method can define the

extent of bleeding in an entire organ (e.g., the cervical canal, in the specific case

analyzed in this paper), which is an improvement over previously described methods

that only define the extent of bleeding in randomly selected tissue sample areas.

When hemorrhage is only measured in a group of representative microscopic areas

(either randomly or deliberately selected), there is a high possibility for error because

there might be areas affected by bleeding that are not included. Using our method, the

bleeding in the entire tissue is analyzed because the Autodesk AutoCAD 2009 program

automatically calculates the surface areas of the tissue affected by bleeding, thereby

preventing a subjective interpretation of the results. Thus, the objectivity of the results

is assured.

The reliability and objectivity of the AutoCAD software was previously proven in

several studies where “polyline” areas were measured in tissue samples and

radiological films.

AutoCAD was used to measure the tomographic bone volume of mandibular ramus

and symphysis bone grafts (Verdugo et al., 2009; Verdugo et al., 2010a). When

compared to direct surgical measurements, the AutoCAD software did not

overestimate the bone volume, indicating that this program could be used to improve

surgical treatment planning prior to sinus augmentation (Verdugo et al., 2010b).

Additionally, AutoCAD was used to quantify necrosis in an experimental model of

acute liver injury (Volarevic et al., 2009) and to estimate the percentage of lung and

liver tissue parenchyma with metastases in a mouse model of breast cancer (Volarevic

et al., 2012). Both studies confirmed that the data obtained with AutoCAD were

precise, highly specific and similar to descriptive histological analysis, confirming the

reliability and accuracy of this software for the quantification of tissue lesions.

In line with these results, we used the Autodesk AutoCAD program to demonstrate that

the hydraulic dilator does not cause significant bleeding during uterine cervix dilation;

no statistically significant difference was found between the bleeding intensities of

tissue samples obtained from patients with undilated uterine cervixes and those from

patients who underwent hydraulic dilation. Using AutoCAD, we calculated the surface

bleeding percentage and confirmed the greater efficacy of the hydraulic dilator with

respect to the damage caused to the cervical tissue. Additionally, the application of

this semi-quantitative method demonstrated that compared to mechanical dilators,

CCBD dilation results in considerably less bleeding in the tissue of the uterine cervix.

It should also be noted that other important and significant advantages of this method

are the simplicity of use of the Autodesk AutoCAD 2009 program as well as its market

availability as a program that is often used for calculations and design in architecture

and engineering.

Conclusion

The semi-quantitative method described in this paper uses the Autodesk AutoCAD

2009 program, which is simple to use and readily available, thereby providing a

novel, objective and precise approach for the estimation of bleeding in tissue samples.

In this study, we determined the efficiencies of different dilators with respect to tissue

bleeding in cervical samples. The greatest level of tissue bleeding was found in

patients whose dilation was performed with a mechanical dilator. The amount of

tissue bleeding in the patients dilated with a hydraulic dilator was similar to that of

patients who were not dilated.

References:

Arsenijevic S, inventor and assignee; Cakic N, inventor and assignee. (2004).

Instrument for fluid injection and dilation probe for implantation in body

cavities, European Patent No.1299146.

Arsenijevic S, Vukcevic-Globarevic G, Volarevic V, Macuzic I, Todorovic P,

Tanaskovic I, Mijailovic M, Raicevic S, Jeremic B. (2012). Continuous

controllable balloon dilation: a novel approach for cervix dilation. Trials. 13,

196.

Bancroft JD, Gamble M. (2008). Theory and practice of histological techniques. 6th

ed. Churchill Livingstone. London.

Biron-Shental T, Fishman A, Fejgin MD. (2004). Medical and mechanical methods

for cervical ripening. Int J Gynaecol Obstet. 85,159–160.

Bokstrom H, Atterfelt P, Alexandersson M, Braannstrom M, Norstrom A. (1998).

Preoperative Cervical Softening Before First Trimester Legal Abortion by

Mifepristone and Misoprostol. Contraception. 58, 157-163.

Gelber S, Sciscione A. (2006). Mechanical methods of cervical ripening and labor

induction. Clin Obstet Gynecol. 49, 642–657.

Greer LA, Millar M, Calder AA. (1992). Gemeprost-induced cervical ripening:

histological and biophysical effects. Eur J Obstet Gynecol Reprod Biol. 47(1),

l-9.

Ichijo M., Suneizu T, Sasai Y. (1976). Histological Aspects of Cervical Ripening.

Tohoku J Exp Med. 118(2), 153-161.

Defrère S, Van Langendonckt A, González Ramos R, Jouret M, Mettlen M, Donnez J.

(2006). Quantification of endometriotic lesions in a murine model by

fluorimetric and morphometric analyses. Hum Reprod. 21(3), 810-7.

Ippolito D, Capraro C, Casiraghi A, Cestari C, Sironi S. (2012 Apr). Quantitative

assessment of tumour associated neovascularisation in patients with liver

cirrhosis and hepatocellular carcinoma: role of dynamic-CT perfusion

imaging. Eur Radiol. 22(4), 803-11.

Kofron CM, Fong VJ, Hoffman-Kim D. (2009). Neurite outgrowth at the interface of

2D and 3D growth environments. J. Neural Eng. 6(1), 016002.

O’Connell K, Jones HE, Lichtenbergb S, Paul M. (2008). Second-trimester surgical

abortion practices: a survey of National Abortion Federation members.

Contraception. 78, 492-499.

Seiji F, Moreira RS, De Angelis MA, Smith Chairman RL. (2009). Orbital asymmetry

in development: an anatomical study. Orbit. 28, 342-6.

Ulin SS, Armstrong TJ, Radwin RG. (1990). Use of computer aided drafting for

analysis and control of posture in manual work. Appl Ergon.. 21, 143-51.

Unal GC, Kaya BU, Taç AG, Keçeci AD. (2009 Apr). A comparison of the efficacy

of conventional and new retreatment instruments to remove gutta-percha in

curved root canals: an ex vivo study. Int Endod J. 42(4), 344-50.

Verdugo F, Simonian K, Smith McDonald R, Nowzari H. (2009). Quantitation of

mandibular ramus volume as a source of bone grafting. Clin Implant Dent

Relat Res. 11, e32-e37. doi: 10.1111/j.1708-8208.2009.00172.x.

Verdugo F, Simonian K, Smith McDonald R, Nowzari H. (2010a). Quantitation of

mandibular symphysis volume as a source of bone grafting. Clin Implant Dent

Relat Res. 12(2), 99-104.

Verdugo F, Simonian K, D'Addona A, Ponton J, Nowzari H. (2010b). Human bone

repair after mandibular symphysis block harvesting: A clinical and

tomographic study . J Periodontol. 81(5), 702-709.

Verdugo F, Simonian K, Raffaelli L, D'Addona A. (2012). Computer-Aided Design

Evaluation of Harvestable Mandibular Bone Volume: A Clinical and

Tomographic Human Study. Clin Implant Dent Relat Res.

doi: 10.1111/cid.12011.

Volarevic V, Milovanovic M, Arsenijevic N, Lukic M. (2009). A new semi-

quantitative method for determining liver damage after concanavalin a

administration. Ser J Exp Clin Res. 11(2), 45-48.

Volarevic A, Ljujic B, Volarevic V, Milovanovic M, Kanjevac T, Lukic A,

Arsenijevic N. (2012). A new semiquantitative method for evaluation of

metastasis progression. J Buon. 17(3), 585-90.

Wang J, Yuan B, Guerrero C, Bahde R, Gupta S, Wang Y. (2011). Quantification of

oxidative DNA lesions in tissues of Long-Evans Cinnamon rats by capillary

high-performance liquid chromatography-tandem mass spectrometry coupled

with stable isotope-dilution method. Anal Chem. 83(6), 2201-9.

Figure legends

Figure 1. The intraepithelial bleeding of the cervical canal epithelium after the

application of a hydraulic dilator (left). A semi-quantitative measurement of the

bleeding was calculated with Autodesk AutoCAD (right).

Figure 2. The subepithelial bleeding of the cervical canal epithelium after the

application of a hydraulic dilator (left). A semi-quantitative measurement of the

bleeding was calculated with Autodesk AutoCAD (right).

Figure 3. The stromal bleeding of the cervical canal epithelium after the application of

a mechanical dilator (left). A semi-quantitative measurement of the bleeding was

calculated with Autodesk AutoCAD (right).

Figure 4. The intraepithelial, subepithelial and stromal bleeding of the cervical canal

epithelium after the application of a mechanical dilator (left). Semi-quantitative

measurements of the bleeding were calculated with Autodesk AutoCAD (right).

Table legends

Table 1. Frequency of patients with bleeding and classification of bleeding in the

tissues.

Table 2. Correlation between the classical pathohistological analysis and our new

semi-quantitative method.

Table 3. Inter-group comparison of the amount of tissue bleeding.

Graph legends

Graph 1. Total area affected by bleeding (in patients with bleeding).

Graph 2. Percentages of the total area affected by bleeding (in patients with bleeding).

Figure 1.

At=A=161.79 Bt=(B1) Bt=(19.20)=19.20

N=Bt x 100/At N=19.20 x 100/161.79 N=11.86%

Figure 2.

At=A=161.79Bt = (B1+B2+B3+B4+B5+B6+B7+B8+B9+B10+B11)Bt=(1.30+1.05+1.45+0.55+0.55+0.42+0.38+0.30+0.45+0.35+0.78)=7.58N=Bt x 100/AtN=7.58 x 100/161.79

N=4.70%

Figure 3.

At=A=161.79Bt=(B1+B2+B3+B4+B5+B6+B7+B8+B9+B10+B11)Bt=(6.35+0.78+0.25+1.13+1.13+0.54+0.22+0.45+0.20+19.00+2.75)=32.80N=Bt x 100/AtN=32.80 x 100/161.79

N=20.30%

Figure 4.

At=A=161.79Bt=(B1+B2+B3+B4+B5+B6+B7+B8+B9)Bt=(3.79+0.63+1.13+

0.19+0.37+0.53+0.33+1.12+0.58)=8.67N=Bt x 100/AtN=8.67 x 100/161.79N=5.35%

Table 1.

Bleeding Group In (%)

Group IIn (%)

Group IIIn (%) Significance

No 33 (82.5%) 4 (10.0%) 27 (67.5%)p = 0.000*

Yes 7 (17.5%) 36 (90.0%) 13 (32.5%)

Classification of bleeding in the tissues n (%)

No bleeding 33 (82.5%) 4 (10.0%) 27 (67.5%) p = 0.000*

group I vs. II:p = 0.000*

group II vs. III:

p = 0.000*

group I vs. II:p = 0.401

Intraepithelial bleeding 6 (15.0%) 4 (10.0%) 9 (22.5%)Subepithelial bleeding 1 (2.5%) 1 (2.5%) 0 (0%)Stromal bleeding 0 (0%) 3 (7.5%) 1 (2.5%)Intra- + subepithelial bleeding 0 (0%) 7 (17.5%) 1 (2.5%)

Intraepithelial + stromal bleeding 0 (0%) 13 (32.5%) 1 (2.5%)

Subepithelial + stromal bleeding 0 (0%) 2 (5.0%) 0 (0%)

Intra-, subepithelial +stromal bleeding 0 (0%) 6 (15.0%) 1 (2.5%)

*Statistically significant.

Table 2.

Intensity of bleeding in the tissuesSurfaces affected by bleeding (patients with bleeding) ρ = 0.542; p = 0.000*

Surfaces affected by bleeding (all patients) ρ = 0.538; p = 0.000*Percentage of bleeding area in the tissue sample (patients with bleeding) ρ = 0.941; p = 0.000*Percentage of bleeding area in the tissue sample (all patients) ρ = 0.941; p = 0.000*

aSpearman correlation; *statistically significant.

Table 3.

Parameters Group IX ± SD;

(Med; min-max)

Group IIX ± SD;

(Med; min-max)

Group IIIX ± SD;

(Med; min-max)

SignificanceAll

three groups

Between groups

Surfaces affected by bleeding (patients with bleeding)

5.63 ± 3.42 (5.22; 1.57-10.61)

30.00 ± 32.91 (14.70; 4.03-123)

7.80 ± 5.77 (6.77; 1.36-24.08)

p = 0.002*

ap = 0.009*bp = 0.521cp = 0.006*

Surfaces affected by bleeding (all patients)

0.84 ± 2.38(0.0; 0.0-10.61)

27.00 ± 32.48(12.45; 0.0-123)

2.53 ± 4.89(0.0; 0.0-24.08)

p = 0.000*

ap = 0.000*bp = 0.056cp = 0.000*

Percentage of bleeding area in the tissue sample (patients with bleeding)

3.48 ± 2.11 (3.22; 0.97-6.55)

12.81 ± 11.18 (9.08; 2.5-41.67)

4.82 ± 3.57 (4.18; 0.84-14.88)

p = 0.003*

ap = 0.009*bp = 0.521cp = 0.007*

Percentage of bleeding area in the tissue sample (all patients)

0.52 ± 1.47(0.0; 0.0-6.55)

11.53 ± 11.28(7.70; 0.0-41.67)

1.57 ± 3.02(0.0; 0.0-14.88)

p = 0.000*

ap = 0.000*bp = 0.056cp = 0.000*

*Statistically significant; agroup I vs. II; bgroup I vs. III; cgroup II vs. III.

Graph 1.

Graph 2.